Anjith George

Diffusion-generated thermal faces can effectively break the multi-modal data bottleneck in biometrics, outperforming models trained on scarce real pairs without requiring costly image translation at inference time.

LoRA adaptation alone can lead to a staggering drop in performance for face presentation attack detection when evaluated across different datasets.

Pretrained vision models can excel in face presentation attack detection, but their performance falters when faced with domain shifts across datasets.

Real-world conditions expose critical performance gaps in face recognition systems, with state-of-the-art models struggling under the challenges of motion blur and occlusion.